Sorry for being the one dropping some hemlock into the cup ... XML will _not_ solve all our problems. It will make some things more manageable and integrateable, but it won't provide any miracles. There's still a need for creative and hard working GIS users and programmers :-)
Compared to simple formats as e.g. CSV based formats, there's always a penalty to pay by utilizing XML, namely the parsing and DOM handling interfaces. The former can be very cpu intensive while the latter hogs huge memory resources. Luckily the prices of both cpu's and memory units are steadily dropping, otherwise XML would be a practical no-go. As for the immidiate practical implementations of GIS relevance : GML is basically flawed by its limitless ambitions and absence of reasonable bounds, making the actual format in severe need of further standardization before it can be called a practical standard for all, imho. Try and ask some of the people who's gotten their hands really "dirty" with GML, and see what they say ;-) SVG is certainly appealing compared to the "dreary" exchange formats we've been using, but when the day ends, it's still "just" a graphics vector format, not a map format. It does have some nice features when it comes to animations etc., but the evenly feature packed proprietary Flash format is still way ahead of SVG both in net penetration and in browser adaptation as I understand it. All that said, I'm all for XML and think it's a nice framework to work with. And in a lot of situations XML can indeed make things better. But only when it truly offers some added value compared to what it replaces. Best regards/Med venlig hilsen Lars V. Nielsen GisPro, Denmark http://www.gispro.dk/ http://www.gispro.biz/ ----- Original Message ----- From: "Neil Havermale" <[EMAIL PROTECTED]> To: "'SCISOFT'" <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]>; "Neil Havermale" <[EMAIL PROTECTED]> Sent: Thursday, March 25, 2004 6:23 PM Subject: MI-L RE: XML and MapInfo > IL - > > > > As a rookie on XML it sounds simple (and increasingly loud)... use XML and > (most of) the GIS data sharing issues goes away...? > > > > And topically to our list and if I get Bill's request right, a XMLed > workspace, wouldn't that be a XLM managing and defining a collection of > XMLs...? Interesting? > > > > I am also curious about your alert regarding the XML vocabulary between the > "<>"s. Is this a normalization issue or a fixed dictionary one? At one > early date in yester years, when both MapX and MapObjects were only 1.0s, > I/we were impressed in how closely either's vocabularies were alike. Not > prefect but enough to suggest a common code vocabulary (Yeah, right, good > luck!)? > > > > That foot note on my experience with assumptions of simplicity aside, I > guess my excitement is lifted with this XML-stuff. It seems to offer > improved methods to ease the endless quest for no-cost and other background > layers. If there is such a benefit to a geoXML outline for information > distribution, meaning easier and less complex IO conversions, better > metadata of all sorts including spatial indices, as well as open > readability, then could not "our" data quest issues evolve to > TerraServer-like XML-automations? How these tabular geoXML lists are then > indexed to the corresponding TAB, SHP, and other geographic objects alerts > my curiosity. I see this XML future as a certainty but for this boy, it > remains yet a poorly understood technology shift. Thanks for your insight > as to where a value might lurk. I will keep watching. > > > > MidNight Mapper > > Aka neil > > > > > > -----Original Message----- > From: SCISOFT [mailto:[EMAIL PROTECTED] > Sent: Wednesday, March 24, 2004 9:48 PM > To: [EMAIL PROTECTED] > Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED] > Subject: XML and MapInfo > > > > Bill / Neil > > > > Bill asked "has anyone here developed an XML DTD or xmlschema to describe a > MapInfo layout?" and "In other words, has anyone attempted to convert a > workspace to XML?" > > > > I'm surprised there hasn't been any cross-talk on this. Surely the .NET > version of MapInfo will be using the GML (XML recommendation for GIS) from > the OpenGIS Consortium? It has been adopted by several vendors. > Particularly, it's used in part by webmap servers. Version 3.0 has been > released and ratified I think. > > > > A couple of MapInfo people have adopted XML for the metadata descriptions. I > guess you know them - I think one is a regular correspondent in MapInfo-L. > > > > The easiest thing of all is to decide on a DTD or XML Schema for the > Metadata and for the Workspace. As with all XML, the first decision is to > agree on the vocabulary to be used (the things between the angled brackets - > like the pair <mapinfo> </mapinfo> ). My guess is that MapInfo Corporation > has long since decided on both of those. > > > > The implementation of the .TAB format in XML is also pretty simple - after > all, it's just a bit of fairly standard text - > > > > !table > > !version 300 > > !charset WindowsLatin1 > > > > Definition Table > > Type NATIVE Charset "WindowsLatin1" > > Fields 11 > > AREA Decimal (16, 3) ; > > PERIMETER Decimal (16, 3) ; > > CITIESX020 Decimal (11, 0) ; > > FEATURE Char (80) ; > > IDNUM Decimal (5, 0) ; > > CAPCODE Decimal (2, 0) ; > > NAME Char (48) ; > > FIPS Char (5) ; > > FIPS55 Char (5) ; > > POP Decimal (11, 0) ; > > STID Char (7) ; > > > > The nice thing about XML is that (for software that supports it) there may > be redundant or not-used definitions within it. Particularly, this applies > to the (new from v6) .WOR workspace that includes printer information - > > > > Set Window FrontWindow() Printer > > Name "HP LaserJet 1100 (MS)" Orientation Portrait Copies 1 > > Papersize 9 > > > > I guess what I am getting around to articulating is that it is necessary for > a user community or software developer to first agree on the vocabulary (tag > names, and what they mean - very precisely) and then it is not a very hard > job to describe the various data structures for the 'components' of > MapInfo's file formats. > > > > You / we / someone should consult with the software vendor / manufacturer > (eg, if it's Bill Thoen who is going to make some software, using MapX for > example, then it is probably to his advantage to consult with others). My > guess is that MapInfo Corporation has decided what it is going to do (though > with XML It is not a hard job to change the terminology / vocabulary, the > data structures may be set in stone already). > > > > There's no point really in developing a suite of MBX for 'older' MapInfo > versions - what for? The area that's crying out for XML-ifying is data > exchange, though - and that has been reasonably addressed by some 3rd-party > data exchange software developers / vendors. The great thing about the use > of XML Schemas is that they are self-describing - the data structure and the > data are explicit within the file (as you would know, not necessarly the > same file, but explicitly by reference URI or by an include, etc). > > > > So, Bill - expand on what you want to do with XML in relation to GIS or > MapInfo files or software. > > > > IL Thomas > > GeoSciSoft - Perth, Australia > > > > --------------------------------------------------------------------- List hosting provided by Directions Magazine | www.directionsmag.com | To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] Message number: 11113
